The overarching goal of this project is to provide new knowledge about the transmission dynamics of multidrug- resistant (MDR) M. tuberculosis in a high tuberculosis-burden area in Peru. A better understanding of the epidemiology and natural history of MDR tuberculosis in a high-exposure population is very significant because it will change our capacity to contain the spread of MDR tuberculosis in the United States and elsewhere. Mutations that lead to antibiotic resistance are often assumed to reduce bacterial fitness in the absence of antibiotics;this is referred to as the fitness "cost" of resistance. Recent studies suggest specific drug-resistance- conferring mutations that minimize the fitness cost of antibiotic resistance may favor the emergence of highly transmissible MDR mutants. The emergence of such M. tuberculosis strains threatens global tuberculosis control because these strains are much more difficult and expensive to treat than drug-sensitive strains. In the 1990s, MDR tuberculosis outbreaks in the U.S. were associated with high mortality. The existence of highly transmissible MDR strains may create an opportunity for the intentional spread of this disease through bioterrorism. The proposed study will build on an established clinical infrastructure in the urban shantytowns in northern Lima, Peru, where tuberculosis is hyperendemic. There, MDR mutants resistant to as many as nine drugs have produced secondary cases. Specific aims are: (1) To measure the within-household transmission of MDR and isoniazid- resistant M. tuberculosis strains compared to drug-sensitive strains. (2) To assess the impact of socio- demographic and clinical confounders and risk modifiers such as age, sex, co-morbidity, HIV infection, nutritional status, and BCG vaccination on these associations. (3) To measure associations among specific resistance mutations and phenotypes including (i) multi-drug resistance, (ii) clinical presentations, and (iii) transmissibility. (4) To measure the rate of true relapse versus re-infection. These aims will be achieved using a prospective four-year enrollment cohort study and molecular epidemiologic methods. Over 11,000 patients, including at least 500 with infectious MDR tuberculosis, will be enrolled along with their exposed household contacts. This project will build on an established collaboration among a multidisciplinary team of U.S. and Peruvian investigators, the Massachusetts Department of Public Health, and the Ministry of Health of Peru;it will permit the enrollment and longitudinal follow-up of a large cohort of persons with sustained exposure to MDR, drug-resistant, and drug- sensitive M. tuberculosis strains. Results will serve to guide basic research in tuberculosis diagnostics and therapeutics.